Justice as a Larger Loyalty

Would it be a good idea to treat 'justice' as the name for loyalty to a certain very large group, the name for our largest current loyalty, rather than the name of something distinct from loyalty? Could we replace the notion of 'justice' with that of loyalty to that group - for example, one's fellow-citizens, or the human species, or all living things? Would anything be lost by this replacement?

Moral philosophers who remain loyal to Kant are likely to think that a lot would be lost. Kantians typically insist that justice springs from reason, and loyalty from sentiment. Only reason, they say, can impose universal and unconditional moral obligations, and our obligation to be just is of this sort. It is on another level from the sort of affectional relations that create loyalty. Juergen Habermas is the most prominent contemporary philosopher to insist on this Kantian way of looking at things: the thinker least willing to blur either the line between reason and sentiment, or the line between universal validity and historical consensus. But contemporary philosophers who depart from Kant, either in the direction of Hume (like Annette Baier), or in the direction of Hegel (like Charles Taylor), or in that of Aristotle (like Alisdair MacIntyre), are not so sure.

Michael Walzer is at the other extreme from Habermas. He is wary of terms like 'reason' and 'universal moral obligation'. The heart of his new book, Thick and Thin , is the claim that we should reject the intuition that Kant took as central: the intuition that 'men and women everywhere begin with some common idea or principle or set of ideas and principles, which they then work up in many different ways'. Walzer thinks that this picture of morality 'starting thin' and 'thickening with age' should be inverted. He says that 'morality is thick from the beginning, culturally integrated, fully resonant, and it reveals itself thinly only on special occasions, when moral language is turned to special purposes.'

Walzer's inversion suggests, though it does not entail, the neo-Humean picture of morality sketched by Annnette Baier in her book Moral Prejudices . On Baier's account, morality starts out not as an obligation, but as a relation of reciprocal trust among a closely knit group, such as a family or clan. To behave morally is to do what comes naturally in your dealing with your parents or children, or your fellow clan-members. It amounts to respecting the trust they place in you. Obligation, as opposed to trust, enters the picture only when your loyalty to a smaller group conflicts with your loyalty to a larger group.

When, for example, the families confederate into tribes, or the tribes into nations, you may feel obliged to do what does not come naturally: to leave your parents in the lurch by going off to fight in the wars, or to rule against your own village in your capacity as a federal administrator or judge. What Kant would describe as the resulting conflict between moral obligation and sentiment, or between reason and sentiment, is, on a non-Kantian account of the matter, a conflict between one set of loyalties and another set of loyalties. The idea of a universal moral obligation to respect human dignity gets replaced by the idea of loyalty to a very large group - the human species. The idea that moral obligation extends beyond that species to an even larger group becomes the idea of loyalty to all those who, like yourself, can experience pain - even the cows and the kangaroos - or perhaps even to all living things, even the trees.

This non-Kantian view of morality can be rephrased as the claim that one's moral identity is determined by the group or groups with which one identifies - the group or groups to which one cannot be disloyal and still like oneself. Moral dilemmas are not, in this view, the result of a conflict between reason and sentiment but between alternative selves, alternative self-descriptions, alternative ways of giving a meaning to one's life. Non-Kantians do not think that we have a central, true, self by virtue of our membership in the human species - a self that responds to the call of reason. They can, instead, agree with Daniel Dennett that a self is a centre of narrative gravity. In non-traditional societies, most people have several such narratives at their disposal, and thus several different moral identities. It is this plurality of identities which accounts for the number and variety of moral dilemmas, moral philosophers, and psychological novels, in such societies.

Walzer's contrast between thick and thin morality is, among other things, a contrast between the detailed and concrete stories you can tell about yourself as a member of a smaller group and the relatively abstract and sketchy story you can tell about yourself as a citizen of the world. You know more about your family than about your village, more about your village than about your nation, more about your nation than about humanity as a whole, more about being human than about simply being a living creature. You are in a better position to decide what differences between individuals are morally relevant when dealing with those whom you can describe thickly, and in a worse position when dealing with those whom you can only describe thinly. This is why, as groups get larger, law has to replace custom, and abstract principles have to replace phronesis  [practical wisdom]. So Kantians are wrong to see phronesis  as a thickening up of thin abstract principles. Plato and Kant were misled by the fact that abstract principles are designed to trump parochial loyalties into thinking that the principles are somehow prior to the loyalties - that the thin is somehow prior to the thick.

Walzer's thick-thin distinction can be aligned with Rawls's contrast between a shared concept of justice and various conflicting conceptions of justice. Rawls sets that contrast as follows:

Šthe concept of justice, applied to an institution, means, say, that the institution makes no arbitrary distinctions between persons in assigning basic rights and duties, and that its rules establish a proper balance between competing claimsŠ [A] conception includes, besides this, principles and criteria for deciding which distinctions are arbitrary and when a balance between competing claims is proper. People can agree on the meaning of justice and still be at odds, since they affirm different principles and standards for deciding these matters.

Phrased in Rawls's terms, Walzer's point is that thick, 'fully resonant' conceptions of justice, complete with distinctions between the people who matter most and the people who matter less, come first. The thin concept, and its maxim 'do not make arbitrary distinctions between moral subjects', is articulated only on special occasions. On those occasions, the thin concept can often be turned against any of the thick conceptions from which it emerged, in the form of critical questions about whether it may not be merely arbitrary to think that certain people matter more than most.

Neither Rawls nor Walzer think, however, that unpacking the thin concept of justice will, by itself, resolve such critical questions by supplying a criterion of arbitrariness. They do not think that we can do what Kant hoped to do - derive solutions to moral dilemmas from the analysis of moral concepts. To put the point in the terminology I am suggesting: we cannot resolve conflicting loyalties by turning away from them all toward something categorically distinct from loyalty: the universal moral obligation to act justly. So we have to drop the Kantian idea that the moral law starts off pure but is always in danger of being contaminated by irrational feelings which introduce arbitrary discriminations among persons. We have to substitute the Hegelian-Marxist idea that the so-called moral law is, at best, a handy abbreviation for a concrete web of social practices. This means dropping Habermas's claim that that his 'discourse ethics' articulates a transcendental presupposition of the use of language, and accepting his critics' claim that it articulates only the customs of contemporary liberal societies.

Richard Rorty is a member of CSD's Council of Advisers. This is an extract from a lecture he gave at the CSD symposium 'An Encounter with Richard Rorty', held on 20 May 1997. The full text appears in Justice and Democracy: Cross-Cultural Perspectives, edited by R. Bontekoe et al (University of Hawai'i Press, 1997).